INDEX

Explanations

punctuation marks indicating a question, excitement, or an ending

oai_token-act-pair · gpt-3.5-turbo

punctuation marks indicating rhetorical questions or exclamations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ibles

-0.75

 anniversary

-0.70

fid

-0.66

clus

-0.64

 bisc

-0.63

 administrator

-0.62

izons

-0.62

 elig

-0.61

aband

-0.61

 scrim

-0.61

POSITIVE LOGITS

 Anyway

1.22

 Similarly

0.99

 Conversely

0.97

 Nevertheless

0.95

 Interestingly

0.91

 Likewise

0.91

 Nonetheless

0.90

 Therefore

0.89

Anyway

0.89

 Finally

0.89

Activations Density 0.116%

punctuation marks indicating a question, excitement, or an ending

punctuation marks indicating rhetorical questions or exclamations

No Comments

No Known Activations